FEATURE TRANSFORMATION FOR EFFICIENTLY IMPROVING PERFORMANCE OF HSC

摘要：

Hyper Surface Classification (HSC) is a novel classification method based on hyper surface which is put forward by Qing He, etc. Experiments show that HSC can efficiently and accurately classify large-size data in two dimensional space and three-dimensional space. Actually, it is difficult to deal with high dimensional data for HSC. So the dimension reduction (data rearrangement) and ensemble methods (feature subspace) are proposed for HSC. But the method based on ensemble will produce many inconsistent and repetitious data in some density dataset, which influence the classification ability of HSC. To solve the problem, a simple and effective kind of data feature transformation method for enhancing performance of HSC is proposed in this paper. The experimental results show that this method can efficiently reduce the inconsistent and repetitious data, efficiently utilize the data Information, and remarkably improve the classification performance of HSC.

关键词： Hyper Surface Classification Ensemble Classification Performance Feature Transformation.

作者: FU-ZHEN ZHUANG QING HE ZHONG-ZHI SHI

作者单位: The Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese The Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese

会议类型: 国际会议

会议名称: 2008 International Conference on Machine Learning and Cybernetics(2008机器学习与控制论国际会议)

会议地点: 昆明

会议语种:英文

页码: 423-428

在线出版日期: 2008-07-12（万方平台首次上网日期，不代表论文的发表时间）